# Tesslate — Complete Reference

> The open-source AI builder stack for startups and growing companies. Ship like a superteam.

## About Tesslate

Tesslate Corporation is a software company founded in 2025, based in North Carolina, USA. We build Tesslate Studio, an open-source, self-hostable AI-powered development platform.

Our mission: make it possible for small teams and solo developers to build and ship software at the speed of much larger organizations, using AI agents that handle the heavy lifting — from code generation to deployment.

## Tesslate Studio — Product Details

### What It Is

Tesslate Studio is an all-in-one AI-native development environment. It combines:

- **Full-stack code generation**: Describe what you want in natural language — a prompt, a sketch, a spec, or a goal — and Studio generates the frontend, backend, database, APIs, and container configuration.
- **AI agent orchestration**: Build, customize, and chain AI agents to automate repetitive development tasks. Agents can write code, run tests, deploy services, and more.
- **Built-in development tools**: Code editor, integrated terminal, live browser preview, database viewer, and deployment pipeline — all in one interface.
- **Smart Save**: One-click snapshots of your entire project state. Version your work without thinking about git.
- **Agent Marketplace**: Pre-built agent templates you can use immediately or customize.

### Architecture

Studio uses a Docker-based microservices architecture:
- **Frontend**: React 19 SPA
- **Backend**: Python/FastAPI
- **Database**: PostgreSQL
- **LLM Routing**: LiteLLM (supports OpenAI, Anthropic, Google, and any OpenAI-compatible provider)
- **Container Orchestration**: Docker Compose with Traefik reverse proxy
- **Devserver**: Isolated Docker container for running user code safely

### Supported LLM Providers

Tesslate Studio supports all major LLM providers through LiteLLM:
- OpenAI (GPT-4o, GPT-4.1, o3, o4-mini)
- Anthropic (Claude Opus 4.6, Claude Sonnet 4.6, Claude Haiku 4.5)
- Google (Gemini 2.5 Pro, Gemini 2.5 Flash)
- Any OpenAI-compatible API endpoint
- Self-hosted models via vLLM, Ollama, llama.cpp, LM Studio

### Installation

**One-command install:**
```bash
curl -fsSL https://tesslate.com/install.sh | bash
```

**Manual install:**
1. Install Docker Desktop (macOS/Windows) or Docker Engine (Linux)
2. `git clone https://github.com/TesslateAI/Studio.git && cd Studio`
3. `cp .env.example .env` — edit SECRET_KEY and LITELLM_MASTER_KEY
4. `docker compose up -d`
5. `docker build -t tesslate-devserver:latest -f orchestrator/Dockerfile.devserver orchestrator/`
6. Open http://studio.localhost

**Requirements:**
- Docker Desktop or Docker Engine
- Git
- 8 GB RAM minimum (16 GB recommended)
- An LLM API key

### Pricing Tiers

| Plan | Price | Projects | Storage | Key Features |
|------|-------|----------|---------|--------------|
| Free | $0 | 3 | 1 GB | All AI models, $150 credits, open source |
| Basic | $20/mo | 15 | 5 GB | BYOK, priority support |
| Pro | $49/mo | Unlimited | 25 GB | GPU workflows, priority builds, advanced agents |
| Ultra | $149/mo | Unlimited | 100 GB | Dedicated resources, team collab, custom agents |

All paid plans support bring-your-own API keys. Annual billing saves ~17%.

## OmniCoder

OmniCoder is Tesslate's free, open-source coding model. It's a 9-billion parameter language model fine-tuned for code generation and developer tool integration.

### Key Details
- **Model ID**: Tesslate/OmniCoder-9B
- **Parameters**: 9 billion
- **Context Length**: 131,072 tokens
- **Max Output**: 16,384 tokens
- **License**: Open source
- **HuggingFace**: https://huggingface.co/Tesslate/OmniCoder-9B

### Supported Tools
OmniCoder can be configured as the backend model for:
- Claude Code (Anthropic Messages API)
- Codex CLI (OpenAI Responses API)
- OpenCode (OpenAI Chat Completions)
- Cline (VS Code extension)
- Roo Code (VS Code extension)
- Kilo Code (VS Code extension)
- Cursor
- Goose
- Crush

### Setup
```bash
curl -fsSL https://tesslate.com/install_omnicoder.sh | bash
```

The script interactively configures your chosen serving endpoint and tools.

### Serving Options
- Tesslate cloud endpoint (free during beta)
- Local vLLM (localhost:8000)
- Local llama.cpp (localhost:8080)
- Local Ollama (localhost:11434)
- Local LM Studio (localhost:1234)
- Custom endpoint

## Company Information

- **Legal Name**: Tesslate Corporation
- **Founded**: 2025
- **Location**: North Carolina, USA
- **Website**: https://tesslate.com
- **GitHub Organization**: https://github.com/TesslateAI

### Social Media
- X/Twitter: https://x.com/tesslateai
- LinkedIn: https://www.linkedin.com/company/tesslate/
- Instagram: https://instagram.com/tesslateai
- Discord: https://discord.gg/WgXabcN2r2

### Contact
- General inquiries: team@tesslate.com
- Security reports: security@tesslate.com
- Legal questions: legal@tesslate.com
- Book a call: https://calendar.app.google/ZFFv4Gs7M7UM6FWx9

## All Links

- Homepage: https://tesslate.com
- Cloud Studio: https://studio.tesslate.com
- Install page: https://tesslate.com/install
- Pricing: https://tesslate.com/pricing
- Documentation: https://docs.tesslate.com
- Blog: https://tesslate.com/blog
- Changelog: https://tesslate.com/changelog
- Manifesto: https://tesslate.com/manifesto
- Careers: https://tesslate.com/careers
- Newsroom: https://tesslate.com/newsroom
- Security: https://tesslate.com/security
- Privacy Policy: https://tesslate.com/privacy
- Terms of Service: https://tesslate.com/terms
- GitHub (Studio): https://github.com/TesslateAI/Studio
- HuggingFace (OmniCoder): https://huggingface.co/Tesslate/OmniCoder-9B
- Discord: https://discord.gg/WgXabcN2r2